Experimental physics is littered with freaky effects, often the product of obscure forces moving and changing objects in ways we don’t expect, but almost always leading to perfectly understandable conclusions. One notable exception – arguably the most notable, in fact – is the double-slit experiment.
Cut two narrow, parallel slits on an opaque sheet and shine light on them. If the conditions are right, you’ll see an interference pattern on the wall behind the sheet – the result, and proof, of the photons’ wavelike behaviour. But if you stick a small detector on each of those slits to track the movement of waves through each one, the interference pattern will be replaced by one or two small pricks of light on the wall – the result, and proof, of the photons behaving like particles and moving only through one slit or the other. Taken together, the experiment demonstrates the wave-particle duality of quantum objects (objects whose behaviour is dictated by quantum forces, as opposed to macroscopic objects that are dominated by classical forces).
In the more than two centuries since the first double-slit experiment, in 1801, many groups of scientists have modified it in different ways to elicit different aspects of the duality, and their implications for the study of the nature of reality. Anil Ananthaswamy’s 2018 book Through Two Doors At Once wends its way through this history, at each step stopping to identify more and more strange discoveries that have only complicated, instead of simplified, the behaviour of particles like photons. One especially weird possibility is contained in a thought-experiment called the Elitzur-Vaidman bomb tester.
Another is contained in a series of famous thought-experiments that American physicist John Wheeler proposed from the late 1970s. Essentially, he asked if each photon could make a ‘decision’ about whether it would travel as a particle or as a wave based on the experimental setup in front of it, if this decision happened in a certain time frame, and if an observer could anticipate this decision-making moment and interfere with it. As bizarre as this sounds, physicists have been able to set up experiments whose results have been consistent with some of Wheeler’s hypotheses.
For example, say you shine a laser at a beam-splitter.

The beam is split in two perpendicular directions; let’s call them A and B. A is made to bounce off a mirror by 90º and moves to a point, which we’ll call P. B is also turned 90º by a mirror in its path and directed to P. If there is a detector at P, physicists have observed a prick of light – indicating both A and B beams were composed of particles. But if there is another beam-splitter at P, then the combined A and B beams are split once again into two beams – and one of them has shown an interference pattern. If A and B were composed of particles until they struck the detector or splitter at P, where did the waves come from? Or, according to Wheeler’s hypothesis, did the photons travelling as part of A and B anticipate that there would be a splitter instead of a detector at P, and decided to become waves? We don’t know. Specifically, there are different interpretations of the experiment’s outcomes that try to make sense of what happened, but we don’t have objective data that supports one exact possibility, in a classical sense.
Wheeler himself concluded that there are no phenomena in the natural universe that are independent of their observations. That is, until you observe something (quantummy) happening, Wheeler figured it wouldn’t have happened (at least not the way you think it did). But more importantly (for this post), both Wheeler’s ideas and the experiments that physicists used to elucidate wave-particle duality kept the focus on the particle, the observer and the test setup. A new study by scientists in the US may complicate this picture even more: they’ve reported evidence that the source of the particles could also influence how the particles behave in an experiment.
Theoretical physicists have anticipated such a finding. For example, one paper published in February 2020 said that when its authors set out to quantify the extent to which a setup would produce an interference pattern or pinpricks of light, they found a simple mathematical relationship between this measure and the purity of the photon source. In the new study, simply put, physicists flashed a specifically tuned laser onto two crystals, of lithium niobate. The crystals then emitted two photons each, which the physicists called the ‘signal’ and the ‘idler’. They directd the signal photons from both crystals to an interferometer – a device that splits a beam of light into two and recombines them to produce an interference pattern – to observe the characteristic proof of wave-like behaviour; they also directed the two idler photons to two detectors, to confirm their particle-like behaviour.
Each pair of signal and idler photons produced by each crystal would be entangled. Wikipedia: “Quantum entanglement is a physical phenomenon that occurs when a group of particles are generated, interact or share spatial proximity in a way such that the quantum state of each particle of the group cannot be described independently of the state of the others, including when the particles are separated by a large distance.” One implication of this relationship is that if we discover, or observe, one of two entangled particles in a certain quantum state, we can determine the state of the other particle without observing it.
In their experiment, the physicists effectively mapped source purity with “the likelihood that a particular crystal source will be the one that emits light” (source). That is, by increasing or decreasing the chances of one of the two crystals emitting photons – by adjusting the strength of the incident laser – the physicists could control the value of the source purity they needed to plug into the equation. They found that when one of the crystals became very likely to emit paired photons, the interference pattern became very feeble – i.e. the photons at the interferometer were behaving like particles. The interference pattern was sharpest when both crystals were equally likely to emit paired photons. These results confirmed the (theoretical) findings of the February 2020 paper, but the physicists were able to do one better.

The February 2020 paper posited that source purity (µ), interference visibility (V) and ‘particle location distinguishability’ (D) were related thus: V2 + D2 = µ2. The new paper also found that µ2 = 1 – E2, where E is a measure of the extent of entanglement between an idler photon and the detector detecting it. This is new, and we don’t yet know how other physicists will exploit it to delve even more into the seemingly bottomless pit that is wave-particle duality. Equally, the experiment also demonstrates, according to Xiaofeng Qian, one of the authors of the February 2020 paper, that a “quantum particle can behave simultaneously, but partially, as both” wave and light.